9 Commits

Author SHA1 Message Date
pablonyx
3a3b2a2f8d add user files (#4152) 2025-04-01 16:19:44 -07:00
evan-danswer
56f8ab927b
Contextual Retrieval (#4029)
* contextual rag implementation

* WIP

* indexing test fix

* workaround for chunking errors, WIP on fixing massive memory cost

* mypy and test fixes

* reformatting

* fixed rebase
2025-03-30 18:49:09 +00:00
joachim-danswer
463340b8a1
Reduce ranking scores for short chunks without actual information (#4098)
* remove title for slack

* initial working code

* simplification

* improvements

* name change to information_content_model

* avoid boost_score > 1.0

* nit

* EL comments and improvements

Improvements:
  - proper import of information content model from cache or HF
  - warm up for information content model

Other:
  - EL PR review comments

* nit

* requirements version update

* fixed docker file

* new home for model_server configs

* default off

* small updates

* YS comments - pt 1

* renaming to chunk_boost & chunk table def

* saving and deleting chunk stats in new table

* saving and updating chunk stats

* improved dict score update

* create columns for individual boost factors

* RK comments

* Update migration

* manual import reordering
2025-03-13 17:35:45 +00:00
pablonyx
20f2b9b2bb
Add image support for search (#4090)
* add support for image search

* quick fix up

* k

* k

* k

* k

* nit

* quick fix for connector tests
2025-03-05 17:44:18 +00:00
Chris Weaver
f25e1e80f6
Add option to not re-index (#4157)
* Add option to not re-index

* Add quantizaton / dimensionality override support

* Fix build / ut
2025-03-03 10:54:11 -08:00
pablonyx
ddec239fef
Improved indexing (#3594)
* nit

* k

* add steps

* main util functions

* functioning fully

* quick nit

* k

* typing fix

* k

* address comments
2025-01-05 23:31:53 +00:00
Richard Kuo
3eb72e5c1d Revert "More efficient Vespa indexing (#3552)"
This reverts commit 27832167819f063bc2f4436765ff27f0ccba2b64.
2024-12-31 09:40:23 -08:00
pablonyx
2783216781
More efficient Vespa indexing (#3552)
---------

Co-authored-by: Chris Weaver <25087905+Weves@users.noreply.github.com>
2024-12-30 18:51:14 -08:00
pablodanswer
21ec5ed795 welcome to onyx 2024-12-13 09:56:10 -08:00